hl-grill
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user-provided files (plans, specs, ideas) and repository files (code, docs, GLOSSARY.md).
- Ingestion points: Reads user-specified input paths, existing repository documentation, and project code in both SKILL.md and references/gaps.md.
- Boundary markers: The instructions do not specify explicit delimiters or warnings to ignore instructions embedded within the files being analyzed.
- Capability inventory: The agent has the capability to write to the filesystem (saving briefs, ADRs, GLOSSARY.md updates, and gaps reports) and invoke subagents for lookups.
- Sanitization: There is no mention of sanitizing or escaping the content of external files before they are interpreted by the agent's logic.
Audit Metadata